RunPaceFlow

Personal running data visualization platform with automatic Strava sync.

中文文档

Features

• Strava / Nike Run Club data import
• Map route visualization with animated playback
• Split pace analysis and charts
• AI-powered running insights (Claude AI)
• Daily auto-sync via GitHub Actions
• Responsive design for desktop and mobile

Configuration

Environment Variables

Create a .env.local file:

# Required - Map style
NEXT_PUBLIC_MAP_STYLE=https://basemaps.cartocdn.com/gl/positron-gl-style/style.json

# Strava (recommended)
STRAVA_CLIENT_ID=your_client_id
STRAVA_CLIENT_SECRET=your_client_secret
STRAVA_REFRESH_TOKEN=your_refresh_token

# Nike Run Club (optional)
NIKE_ACCESS_TOKEN=your_access_token

# Claude AI (optional - for running insights)
ANTHROPIC_API_KEY=your_api_key
ANTHROPIC_BASE_URL=  # Optional: custom API endpoint for proxy

# Goals (optional - customize weekly/monthly targets)
NEXT_PUBLIC_WEEKLY_DISTANCE_GOAL=10000   # Weekly distance goal in meters (default: 10km)
NEXT_PUBLIC_MONTHLY_DISTANCE_GOAL=50000  # Monthly distance goal in meters (default: 50km)
NEXT_PUBLIC_WEEKLY_DURATION_GOAL=3600    # Weekly duration goal in seconds (default: 1 hour)
NEXT_PUBLIC_MONTHLY_DURATION_GOAL=18000  # Monthly duration goal in seconds (default: 5 hours)

Getting Strava Token

1. Go to Strava API Settings and create an app
2. Get your Client ID and Client Secret
3. Obtain Refresh Token via OAuth flow (see strava-oauth guide)

Claude AI Setup (Optional)

The AI feature generates personalized running insights for each activity, analyzing your pace, splits, and performance patterns.

For Vercel Deployment:

Configure these environment variables in Vercel Dashboard (Settings → Environment Variables):

Variable	Description
ANTHROPIC_API_KEY	API key from Anthropic Console
ANTHROPIC_BASE_URL	(Optional) Custom API endpoint for proxy

For Local Development:

Add to your .env.local file:

ANTHROPIC_API_KEY=your_api_key
ANTHROPIC_BASE_URL=  # Optional: custom API endpoint

Note: Without Claude AI configured, the app works normally but without AI insights.

Local Development

# Install dependencies
bun install

# Initialize database
bun run db:push

# Start dev server
bun run dev

# Manual sync
bun run sync

Visit http://localhost:3000

Deployment

Option 1: Vercel (Recommended)

1. Fork this repository
2. Import project in Vercel
3. Configure environment variables (same as .env.local)
4. Deploy

Note: Vercel deployment requires GitHub Actions for data sync. Database is stored in the repository.

Option 2: Docker (WIP)

# Using docker-compose
docker compose up -d

# Or manual build
docker build -t runpaceflow .
docker run -d -p 3000:3000 \
  -e NEXT_PUBLIC_MAP_STYLE=https://basemaps.cartocdn.com/gl/positron-gl-style/style.json \
  -e STRAVA_CLIENT_ID=your_id \
  -e STRAVA_CLIENT_SECRET=your_secret \
  -e STRAVA_REFRESH_TOKEN=your_token \
  -v runpaceflow_data:/app/data \
  runpaceflow

GitHub Actions Auto-Sync

Built-in GitHub Actions workflow for daily activity sync.

Configure Secrets

Add in repository Settings → Secrets and variables → Actions:

Secret	Description
STRAVA_CLIENT_ID	Strava Client ID
STRAVA_CLIENT_SECRET	Strava Client Secret
STRAVA_REFRESH_TOKEN	Strava Refresh Token
PAT	Personal Access Token for push access

Sync Schedule

• Auto sync: Daily at UTC 0:00 (Beijing 8:00) Customize by editing cron in .github/workflows/sync.yml:

  on:
    schedule:
      - cron: '0 0 * * *' # UTC time, format: min hour day month weekday

• Manual trigger: Actions → Sync Activities → Run workflow
• Data storage: SQLite database auto-committed to data/activities.db

Credits

Inspired by yihong0618/running_page

License

MIT